Whether you scaled globally first and figured out MarTech and data “later,” or you’ve grown through rapid M&A and inherited a patchwork of regional stacks, the challenge is the same: too many clouds, too many tools, too many versions of the customer. For enterprise leaders, the mandate is now clear—consolidate the complexity, reduce total cost of ownership, and build a governed foundation for compliant growth at global scale.
Why This Problem Is Showing Up Now
Many businesses grew up too fast, or grew through organic M&A’s. Regardless how your business got there, you’re now global and the business did not design its current marketing, data and identity ecosystems—they inherited them or it was an afterthought of legacy thinking and priorities.
Recent research reinforces this urgency: many organizations have invested heavily in Martech yet still “struggle to clearly articulate or measure ROI,” often because fragmented stacks, poor user adoption; which speaks to lack of strategy and prioritized execution roadmap; and poor integration preventing technology from delivering meaningful business impact at scale.
McKinsey notes, unlocking the next wave of growth requires reimagining Martech not as disconnected tools, but as an integrated, enterprise-wide system aligned around the customer. It’s shift from a sunk-cost POV to understanding how MarTech enables and contributes enterprise value, and an engine for revenue growth. See McKinsey article Rewiring MarTech: From Cost Center to Revenue Growth.
Why this problem is surfacing now, lets briefly touch on where it all started.
Typically with enterprises, over time, regions moved off-premise into the cloud, and setup different cloud platforms (AWS, Azure, GCP, OCI), data warehouses (Snowflake, Databricks, BigQuery), and marketing tools (Adobe, Oracle, SalesForce or a hybrid toolbox that make up your MarTech) stack.
What once enabled regional speed now creates enterprise drag: rising costs, duplications, conflicting customer identities and records, inconsistent consent enforcement, reporting disputes, and difficulty scaling globally.
Executives and leaders are now tasked with driving down the cost-to-serve (OPEX), improve the ROCI, etc. all through consolidating, simplifying, and scaling—while reducing TCO (total cost of ownership).
This is not a tool replacement exercise. It is an architectural realignment.
Start at the Root Problem
Most organizations confuse where data is stored with how customers are understood. Let alone, rarely had the luxury to define strategy informed by internal and external user-stories. To to set the stage to understand the root problem, view your architecture like this:
- Your Land = Your cloud platforms. Think Amazon (AWS), Microsoft (Azure) or Google (GCP)
- Your Home & Buildings = Data platforms are the tools and tech like Snowflake, Databricks, and BigQuery where your data is organized, structured, and made usable.
- Your Car = Marketing platforms — the delivery vehicles that move your message, carry the customer experience, and take you to the right destinations across channels (email, SMS, web, mobile, ads).
- Your City Rules = Governance + the global data model — the shared standards (customer ID, consent, definitions, naming) that turn regional chaos into one coordinated system so every team builds and operates off the same “truth.”
The missing layer is harmonization and identity governance.
Architectural Models: Centralized or Federated?
When you’re consolidating a global MarTech stack, there are two practical ways to organize the mess: bring everything into one shared “global core,” or let each region keep its setup while following the same rules and avoid the politics ;-).
It’s important we touch on this point as it will help explain what they offer, and what they cost you over time.
The way I would explain this, it’s the difference between running one well-managed headquarters for the whole company versus running multiple regional offices that follow the same playbook—but can drift without super strong oversight and governance (and when does that often work seamlessly?)
Model A – Centralized Global Harmonization:
All regional data converges into one governed harmonization layer before activation.
Pros:
- Single source of truth
- Strong compliance
- Unified reporting,
- Lower long-term TCO.
Cons:
- Higher initial integration effort
- Higher organizational alignment required.
Model B – Federated Harmonization (Mirrored Across Regions):
The federated approach, or mirror or replication approach. Each region applies the global standard schema locally.
The global standard typically includes:
- Universal Customer ID framework — how identities are created, matched, and maintained across systems.
- Consent model — standardized fields for opt-in status, source, jurisdiction, timestamp, and permitted channels.
- Core data definitions — consistent meaning for key attributes like customer status, account type, lifecycle stage, and engagement signals.
- Product and service taxonomy — shared naming and categorization so offerings are understood the same way globally.
- Channel eligibility rules — how contactability is defined (emailable, marketable, suppressions, etc.).
- Regulatory classifications — flags that align with CASL, GDPR, CCPA, PIPL, and other regional requirements.
- Data quality and formatting standards — normalization of fields such as country codes, addresses, and identifiers.
- Naming conventions and metadata structure — ensuring data can be aggregated and reported consistently.
Ok…your cheat notes, the good and the ugly of the federated model.
Pros:
- Faster deployment
- Regional autonomy.
Cons:
- Governance drift
- Identity duplication
- Higher long-term operational costs.
The tradeoff
View it like this. Model B or the federated approach optimizes for speed today.
Model A or central approach optimizes for scale tomorrow, but takes more time to get there.
So what’s in a global data model then?
I’m usually a red pill person here—if I’m starting with a blank canvas and no major technical debt or politics that would hinder balancing speed with value realization—I lean to the global centralized model. Here’s why…
The global data model defines how the enterprise understands customers EVERYWHERE.
It includes:
- Universal customer ID logic
- Consent attributes by jurisdiction
- Product taxonomy
- Lifecycle status definitions
- Channel eligibility rules
- Regulatory classifications
This model must be centrally governed, even if regions retain separate cloud platforms.
So if you take this path, then the natural fork in the road will be how you choose to handle the data to build your Customer 360 or universal customer identity and their profile.
Customer 360 & The Role of a CDP
A CDP (Customer Data Platform) such as SalesForce Data Cloud, Adobe Real-time CDP or Segment (Twilio) serves as the engine for customer identity and consent.
Picture this. You have 5 different tools or systems where a single customer lives. Let’s call her Lisa. Her profile and data lives in many siloed or fragmented systems in your business:
- Marketing automation tool: Lisa receives emails, SMS and push notifications from marketing
- Website: Lisa visits your brand website to stay connected and see what’s new.
- eCommerce: Lisa browses and shops and transacts in your eStore.
- POS: Lisa sometimes like to go offline to your Robson street store to purchase in-store.
- CRM: Your sales reps. created a record for her under the company account she works for.
- Social: Lisa stays engaged and connected across Meta, Instagram and TikTok
- 3rd Parties: Other data providers may have data about her, whether its her local bank, other stores or competitors she shops at, tickets to a concert she attended, etc.
So now that Lisa’s data dna is everywhere, thus fragmented and siloed, imagine what you could do if you had one simple universal or unified identity and profile of Lisa from all the data across those systems
Think hyper-personalization and experiences you could enable to keep her engaged with your brand and your eco-system for future up-sell, cross-sell and long-term loyalty.
So unifying all the profile data about Lisa, reconcile her consent, creates activation-ready audiences like Lisa and bridges the enterprise data about Lisa and all your customers into marketing platforms for orchestration and activation.
A CDP is by no means a replacement for a data warehouse. It sits between enterprise data and marketing orchestration and activation.
Where Should Consent Live?
Consent should live above marketing tools.
Enterprise systems store consent events.
The CDP resolves and governs consent.
Marketing platforms execute only on approved audiences.
This structure reduces regulatory risk and ensures global enforcement consistency across CASL, GDPR, CCPA, PDPA or mixes of them.
Marketing Orchestration Layers: B2C vs B2B
One of the most common MarTech drivers of architecture discussion and needs is driven by marketing automation, personalization and its orchestration.
While the best-of-breed tools from SalesForce, Adobe, Oracle are all great enterprise tools, the reality is most of you have these plus a dozen or two of other mid-tier MarTechs.
But in spirit of understanding the Marketing activation layer, lets look at one of the most effective MarTech solutions; marketing automation.
Let’s take SalesForce for a second (not a paid plug).
SalesForce, their B2C Marketing (Marketing Cloud Engagement) focuses on lifecycle marketing at scale, multichannel orchestration, and behavioral engagement. For those of you not familiar with lifecycle marketing, just think of it as all those high-value touchpoints when engaging a business; from staying top-of-mind, nurturing, converting, welcoming, onboarding, up-sell, cross-sell, retention and loyalty opportunities to touch the customer– and do it at scale and hyper-personalized and tailor to the individual (across their channel preference).
SalesForce B2B solution, SalesForce Marketing Account Engagement focuses is tailored to the B2B crowd where lead nurturing, qualification, scoring, pipeline management, CRM alignment, and sales pipeline acceleration are the priorities.
Their marketing automation solutions serve different needs and different orchestration purposes but rely on the same unified identity foundation.
So whether its marketing automation, journey orchestration, segmentation, website personalization, customer profiling or BI reporting, a unified identity and profile are key.
The Outcome: Solve the Original Problem
By centralizing data harmonization and user identity before Marketing activation, businesses like yours reduce duplication, improve compliance, improve customer consent, lower TCO, and enable scalable personalization.
The enterprise shifts from fragmented regional silos to coordinated global orchestration with local execution.
To wrap it up and keep it simple, think of your enterprise architecture blueprint for global marketing and Martech like this:
- Regional Tools: CRM, eCommerce, Marketing Automation
↓ - Regional Cloud Systems (AWS / Azure / GCP)
↓ - Data Platforms (Snowflake / Databricks / BigQuery)
↓ - Global Harmonization Layer (Canonical Data Model)
↓ - CDP / Identity Resolution
(ex. 1 identity for a customer across all systems and tools, stitching or unifying 1st party, 2nd party and 3rd party data)
↓ - Marketing Activation (B2B or B2C)
- Marketing automation
(email, SMS, push, digital advertising, website) - Journey orchestration
- Personalization
- CMS and content fragments
- eCommerce platform
- Website Analytics
(enabled with the latter)
- Marketing automation
Got questions or ideas to share on your journey to unify your MarTech stack and realize that one single customer identity? Reach out to schedule a call or connect on LinkedIn.